Shared Care Records (SCRs) have undergone significant evolution over the past two decades. What began as relatively simple mechanisms for sharing documents across organisational boundaries is now moving towards intelligent, conversational systems that actively support clinicians, patients, and health systems. This evolution can be understood in four broad stages, each building on the limitations of the previous one.
Introduction
Over the past 15 years, we have had the opportunity to work at the forefront of Shared Care Record programmes across multiple countries and health systems. From our first national deployments in Slovenia and Malta, to our recent deployments for cities like London and countries like Greece and Ireland, we have seen first-hand how these initiatives have evolved. From early, document-based exchanges designed primarily for visibility, through to today’s large-scale, ecosystem-driven deployments that actively support clinical care, coordination, and population health.
Our journey closely mirrors the broader maturation of Shared Care Records as a concept. We started with implementations focused on sharing clinical documents across organisational boundaries, addressing the most basic problem of information silos. Over time, these programmes progressed towards structured, longitudinal records capable of supporting collaboration across primary, secondary, community, and social care. More recently, we have been involved in deploying SCRs as national platforms, enabling multiple applications and services to operate on top of a common, vendor-neutral data foundation.
The most recent example of this evolution is the national Shared Care Record for the Health Service Executive (HSE) in Ireland. This programme represents a culmination of lessons learned over more than a decade: moving beyond documents, embracing structured data at scale, and deliberately designing the SCR as an enabling platform for future applications, analytics, and AI-driven services.
This accumulated experience provides a practical lens through which to understand the evolution of Shared Care Records. The four stages outlined below are not theoretical, they reflect real deployments, real constraints, and real progress observed across national and regional programmes as SCRs transition from passive repositories into active participants in care delivery.
Document-based exchange: Digitising and sharing information
The first generation of Shared Care Records focused on document- based exchange. The primary goal was to make clinical documents, such as discharge summaries, clinical notes, and test reports, available across all care settings. Technically, this was often achieved through document repositories, portals, or health information exchange (HIE) infrastructures.
This stage delivered a vital breakthrough: clinicians could see information created elsewhere. However, these systems were essentially read-only. While documents could be uploaded and viewed, they could not easily be structured or enriched once shared. Crucially, third-party applications and clinical systems were typically unable to contribute new data directly into the shared record in a computable way.
In practice, this meant SCRs functioned as passive archives rather than living records. Data was locked inside PDFs, scanned documents, or proprietary formats. Clinicians still had to re-enter information into their local systems, and innovative applications, such as decision support tools, care coordination apps, and analytics, could not update the record. The result was limited clinical impact and minimal workflow integration.
Structured shared records: From documents to data
The second stage transitioned beyond documents to structured, longitudinal, shared records. Instead of exchanging files, systems began sharing granular clinical data: problems, medications, allergies, observations, encounters, and plans.
This shift enabled a single, coherent patient record to be assembled across multiple care settings. Data could be queried, filtered, and reused, rather than merely read. Clinicians gained faster access to relevant information, thus reducing duplication of tests and assessments. Due to the structured nature of data, alerts and notifications became possible.
Importantly, this stage allowed multiple systems to both consume and contribute data. The shared record became a collaboration layer rather than a static repository. Examples include national or regional shared records that aggregate data from hospitals, primary care, community services, and social care, forming a continuously updated longitudinal view of the patient.
While powerful, these systems were still largely system-centric. Interaction models remained based on screens, forms, and predefined workflows, and the cognitive burden on clinicians remained high.
Ecosystems on top of the SCR: Interoperability as a platform
In the third stage, the Shared Care Record becomes a digital health platform rather than just an integration artefact. Once high-quality structured data is available, it can support an ecosystem of applications layered on top of the shared record.
At this point, standards such as the International Patient Summary enable consistent cross-border and cross-vendor data exchange. At the same time, APIs allow specialised applications to plug into the shared data layer. Clinical use cases expand beyond viewing data to actively supporting care processes.
Real-world examples illustrate this shift clearly. The Universal Care Plan (UCP) in London demonstrates how a shared record can support multidisciplinary care planning around a single, patient-centred plan. In Slovenia, eKarton demonstrates how national shared data can underpin a growing array of clinical and administrative services, facilitating care coordination for the prevention and management of chronic diseases. Greece’s national EHR similarly builds on shared data foundations to enable both professional and citizen access.
At this stage, SCRs support not just clinicians, but also patients. However, while ecosystems of apps are possible, they still depend on humans to interpret information and take action.
Conversational and agentic records: From access to assistance
The fourth stage represents a qualitative shift: the Shared Care Record becomes conversational, proactive, and agent-enabled, allowing users to interact with the record.
Instead of navigating screens and forms, clinicians and patients can interact with the record using natural language. Under the hood, protocols such as the Model Context Protocol (MCP) enable AI systems to access structured clinical data safely and consistently. The result is not simply “chat with your EHR,” but context-aware interaction grounded in the patient’s longitudinal record.
In this model, the SCR does more than respond to queries. Intelligent agents can:
- Proactively identify care gaps (e.g. missing reviews, overdue tests, guideline deviations).
- Surface relevant information at the point of care.
- Support clinical reasoning and shared decision-making.
- Assist patients in understanding and managing their conditions.
- Enable secondary uses of data for research and population health, aligned with frameworks such as the European Health Data Space.
In Slovenia, eKarton demonstrates how national shared data can underpin a growing array of clinical and administrative services, facilitating care coordination for the prevention and management of chronic diseases.
Critically, this stage extends the application ecosystem further, encompassing not just apps, but also agents that operate on top of the shared record, each focused on specific tasks or domains. The Shared Care Record becomes an active participant in care delivery, not just an information source.
Notably, Greece already enables patients to prompt their own health records using natural language, lowering barriers to access and engagement.
In summary
Shared Care Records have evolved from basic document-sharing solutions into a core component of health-system infrastructure. Early document-based approaches improved visibility across organisations but were essentially read-only and poorly integrated into clinical workflows. The move to structured, longitudinal shared records enabled true collaboration, allowing multiple systems to contribute to and reuse clinical data. More recently, SCRs have been designed as platforms that support ecosystems of applications for care planning, coordination, analytics, and patient access without relying on monolithic systems.
Shared care records have been designed as platforms that support ecosystems of applications for care planning, coordination, analytics, and patient access without relying on monolithic systems.
The next stage further extends this, introducing conversational and agent-based capabilities. These enable Shared Care Records to transition from passive access to active assistance, proactively identifying care gaps and supporting informed decision-making. Overall, SCRs are progressing from documents to data, from visibility to collaboration, and from access to assistance, reflecting their growing role in delivering coordinated, sustainable care at scale.
At Better, our digital health platform has been shaped by more than ten real-world regional and national Shared Care Record deployments, operating at the scale, complexity, and governance levels required by modern health systems. We have progressed through every architectural phase from document exchange and HIE integration to the persistence of a longitudinal shared care record, and ultimately, to an ecosystem platform approach. Central to this evolution has been our uncompromising commitment to open standards, including IHE, HL7, and openEHR, ensuring vendor neutrality, semantic interoperability, and long-term regulatory alignment.
This foundation enables national- scale scalability, supporting tens of millions of patient records within a single longitudinal data layer, while meeting the requirements of cross-border data sharing, secondary use, and data-space governance anticipated under the European Health Data Space. Crucially, the same architecture makes the platform inherently AI-ready: structured, high-fidelity longitudinal data supports advanced analytics, clinical decision support, and next-generation AI and agentic workflows, without re-platforming or loss of clinical meaning.
This blog is part of our Mission 2026 edition. If you want to read the whole Mission, dedicated to Shared Care Records, and all the work Better has been doing all over the world, you can do so below.
